Intel and Foxconn announced a strategic collaboration at Computex 2026 in Taipei to build next generation AI infrastructure and intelligent computing platforms, giving both companies a new push into the hardware layer behind AI deployment.

The partnership combines Intel’s processor architecture, silicon technologies, and software ecosystem with Foxconn’s manufacturing scale and system integration capabilities. 

The companies said the collaboration is aimed at meeting rising demand for AI computing systems across data centers, edge computing, and physical AI applications.

The project also includes SambaNova, whose AI platform is being combined with Intel Xeon processors in rackscale infrastructure designed for hyperscale deployments. 

Intel said the production ready racks will pair Xeon processors with SambaNova SN 50 reconfigurable dataflow units to support AI inference with better cost and power efficiency.

The collaboration targets several bottlenecks in AI data centers, including high speed interconnects, system telemetry, cooling systems, and energy efficiency. 

Intel is bringing advanced chip and interconnect technologies, while Foxconn is expected to turn those components into integrated systems that can be manufactured and deployed at scale.

Foxconn also plans to manufacture a CPU dense version of the rackscale infrastructure for workloads that do not require additional acceleration. That system is designed for cost optimized inference, data processing, and hybrid AI workloads.

The companies are also looking beyond traditional data centers. The partnership extends to edge AI and physical AI use cases, including robotics, automotive systems, smart city infrastructure, industrial computing, and smart manufacturing.